Intelligent duck farm environment management optimization method and system

Abstract

The invention relates to the technical field of livestock and poultry breeding, in particular to an intelligent duck farm environment management optimization method and system, and the method comprises the steps: collecting real-time growth state information of breeding duck growth, employing a sensor to optimally collect environment influence factors concerning a breeding duck farm, and executing association to obtain the growth state information after information synchronization; filtering high-weight environment influence factors based on growth attributes in the growth state information blocks, and setting key environment influence factors of a plurality of growth scenes based on the high-weight environment influence factors; environment influence factor collection of a plurality of growth scenes is executed, a matching environment log of healthy growth of the breeding ducks is configured to execute semantic processing, and the similarity with the matching environment log is collected; and optimizing the breeding duck growth scene based on the similarity to obtain priority optimization entries, and optimizing the environment of the concerned duck farm. According to the method, the optimal value can be judged according to the duck breeding state data of the intelligent duck farm and the mixed breeding state of the breeding ducks, and the utilization efficiency and the output result of the duck farm are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of livestock and poultry breeding technology, and in particular to an intelligent duck farm environmental management optimization method and system. Background Art

[0002] In my country's breeding duck production, more than 80% of duck products come from a small number of duck farms and free-range farmers, which have the problems of low productivity, short production cycle and low breeding profits. Moreover, due to the dual pressures of the market and epidemics, the profits of breeding ducks are getting lower and lower.

[0003] Moreover, my country's poultry breeding industry still has certain problems in the production process, mainly manifested in: unreasonable layout, simple infrastructure, aging equipment, outdated breeding concepts; high breeding density in the region, serious manure problems, exceeding the regional environmental safety carrying capacity, and many safety hazards for breeding ducks, etc. Therefore, the development of a scientific and standardized intelligent duck farm environmental management optimization method is crucial to improve the economic benefits of duck breeding. Summary of the invention

[0004] To achieve the above objectives, this application provides the following technical solutions: According to a first aspect of the present invention, the present invention claims protection for an intelligent duck farm environment management optimization method, comprising the following steps: Collect the real-time growth status information of the breeding ducks, use sensors to optimize the collection of environmental factors affecting the duck farm, and associate the historical environmental factors with the growth status information to obtain the growth status information after information synchronization; Clustering the synchronized growth status information through multiple growth scenarios, filtering high-weight environmental influencing factors based on growth attributes in multiple growth status information blocks, and setting key environmental influencing factors of multiple growth scenarios based on the high-weight environmental influencing factors; The key environmental influencing factors are used to collect environmental influencing factors of multiple growth scenarios, and matching environmental logs for healthy growth of breeding ducks are configured, semantic processing is performed on the collected environmental influencing factor logs, and the similarity between the collected and matching environmental logs is calculated; Based on the similarity, the growth scenario of the breeding ducks is optimized to obtain multiple priority optimization items, and the optimization items are used to perform environmental optimization of the duck farm.

[0005] Furthermore, the mixed breeding status of breed ducks of each duck farm is collected according to the duck farm management server, and the growth progress deviation value of each breed duck is analyzed according to the mixed breeding status of breed ducks; If the growth progress deviation value of the breeder ducks is greater than the normal growth progress deviation value, collecting the real-time video mixed breeding status of the duck farm, and collecting the video recognition result based on the real-time video mixed breeding status of the duck farm; Collecting the key growth time nodes of each breed of ducks, and generating the market value effect of each duck farm according to the video recognition results and the key growth time nodes of each breed of ducks; The rearing environment of each duck farm is adjusted based on the market value of each duck farm, and the quality status of the feed in the duck farm is collected, and prompts are given according to the quality status of the feed in the duck farm.

[0006] Furthermore, the mixed breeding status of breed ducks of each duck farm is collected according to the duck farm management server, and the growth progress deviation value of each breed duck is analyzed according to the mixed breeding status of breed ducks, specifically: The low-value mixed breeding status in the duck farm is collected according to the historical breeding records, and the species information of the breed ducks in the duck farm is determined according to the low-value mixed breeding status in the duck farm, and the query tag is constructed based on the species information of the breed ducks in the duck farm; The breed type of each breed of duck is collected according to the query tag, and the relevant environmental type monitoring equipment is deployed according to the breed type of each breed of duck, and a breeding monitoring platform is constructed based on the relevant environmental type monitoring equipment; The mixed breeding status of breed ducks in each duck farm is collected through the breeding monitoring platform, and the matching breed duck data of each breed duck is collected according to the historical breeding records to build a growth progress detection model for breed ducks; The growth prospects of breed ducks are preset, and the mixed breeding status of breed ducks in each duck farm, the matching breed duck data of each breed duck, and the growth prospects of breed ducks are input into the growth progress detection model of breed ducks to generate the growth progress deviation value of each breed duck in each duck farm.

[0007] Furthermore, based on the historical environmental influencing factors and the growth status information, the growth status information after information synchronization is obtained, specifically: Collecting the growth information of the breeding ducks of the duck farm concerned within a preset period, parsing the growth information of the breeding ducks, collecting key scenes of the growth of the breeding ducks, clustering the parsed growth information of the breeding ducks, and collecting the growth information set of the key scenes based on the clustering records; Based on the growth information set collection process characteristics, query the growth samples of breeding ducks by configuring the query ID through the process characteristics, collect the growth sample records of breeding ducks that meet the preset quality based on similarity analysis, and merge and collect the breeding strategies of real-time growth means for multiple breeding duck growth samples; Creating real-time growth status information of the breeding ducks based on the growth information set obtained by clustering driven by the breeding strategy, and performing expansion of the growth status information based on the synchronization of the breeding ducks growth information; Through the GPS positioning system combined with wireless sensor elements, the environmental data block optimization of the duck farm is carried out to optimize the environmental influencing factors of multiple partitions and collect historical environmental influencing factors within a preset period; The historical environment influencing factors are associated with the growth status information, the information influencing factors are set based on the growth status value in the growth status information, and the historical environment optimization information is fused with the growth status information through the information influencing factors to obtain the growth status information after information synchronization.

[0008] Furthermore, high-weighted environmental influencing factors are filtered based on growth attributes in multiple growth status information blocks, and key environmental influencing factors of multiple growth scenarios are set based on the high-weighted environmental influencing factors, specifically: By executing a preset cluster number in the synchronized growth state information of the key scene, a neighborhood range is preset in multiple first cluster number scenes, the growth state value in the neighborhood range is searched, and the scene edge is set based on the mean value corresponding to the growth state value of each cluster after the preset clustering; The loss function is set by cosine distance, and the cosine distance of the growth state value in the neighborhood range from the edge of the two adjacent scenes is determined. When any cosine distance is not less than the preset distance upper limit, the growth state value is associated with the growth scene with high loss. When both cosine distances are not greater than the preset distance upper limit, the growth state value is sequentially associated with the two adjacent growth scenes. When the clustering of the growth state values ​​in the neighborhood range of all the first cluster number scenes is completed, the cluster record of the synchronized growth state information is collected; Collecting growth attributes based on growth status values ​​in multiple growth status information blocks, analyzing linear correlation coefficients in environmental influencing factors through the growth attributes, and filtering environmental influencing factors of multiple growth status information blocks using the linear correlation coefficients; A breeding duck productivity detection model is configured, the breeding duck productivity detection model is trained by environmental influencing factors of multiple growth status information blocks, corresponding detection confidences are collected, maximizing the detection confidences is focused on, and the weights of environmental influencing factors are evaluated through control experiments; The environmental influencing factors of the plurality of growth status information blocks are sorted by weight, and a preset number of high-weight environmental influencing factors are found and set as key environmental influencing factors.

[0009] Furthermore, the key environmental influencing factors are used to collect environmental influencing factors of multiple growth scenarios, and a matching environmental log for the healthy growth of breeding ducks is configured, specifically: Collecting key environmental influencing factors corresponding to a plurality of growth status information blocks, collecting influencing factors from historical environmental influencing factors based on the key environmental influencing factors, and filtering influencing factors based on corresponding breeding duck productivity among the collected influencing factors; Collect the number of influencing factors under key environmental influencing factors, filter the execution number of key environmental influencing factors that are not greater than the preset number, import the influencing factors corresponding to the numbered key environmental influencing factors into the servo network to collect candidate production states, and adopt the intrusion strategy; The sensor structure of the servo network is set to the candidate production state of the server learning influencing factors, the growth information of the breeding ducks of the influencing factors is collected and associated with the same scene, and a listener is added to the candidate production state layer to prompt the growth information of the breeding ducks with too large distribution similarity of the candidate production state patterns; The influencing factors corresponding to the numbered key environmental influencing factors are expanded through iterative training of the intrusion strategy. When the number of influencing factors meets the preset number, the iteration is stopped, and the influencing factors corresponding to the expanded numbered key environmental influencing factors are output; The matching environment logs for the healthy growth of breeder ducks in multiple growth scenarios are configured through influencing factors after filtering and information expansion.

[0010] Furthermore, semantic processing is performed on the collected environmental influencing factor logs to collect the similarity with the matching environmental logs, specifically: Based on key environmental influencing factors, environmental influencing factors of multiple growth scenarios are collected, and environmental influencing factor logs of multiple growth scenarios are obtained in combination with the optimized time; Based on the historical average breeding duck productivity, the preset weights of multiple growth scenarios are obtained, and based on the historical breeding duck growth information, the values ​​and emergency samples of multiple growth scenarios are collected to analyze the emergency rates of multiple growth scenarios; Determine the semantic clustering levels of multiple growth scenarios by using the preset weights in combination with the emergency rate, perform clustering of environmental influencing factor logs based on the semantic clustering levels of multiple growth scenarios, and collect corresponding environmental influencing factor semantic logs; The matching environment log is called to obtain the corresponding matching environment semantic log, the average distance between the environment influencing factor semantic log and the matching environment semantic log is analyzed, and the similarity is obtained based on the average distance log.

[0011] Furthermore, based on the similarity, the optimization of the breeding duck growth scene is performed to obtain multiple priority optimization items, and the optimization items are used to perform the environmental optimization of the duck farm, specifically: Preset similarity upper limit range, associate optimization priorities for multiple preset similarity upper limit ranges, determine the corresponding optimization priority for the similarity upper limit range in which the collected similarities fall, and obtain the optimization items to be transmitted and recorded in a list according to the preset method; When the optimization priority is greater than the preset priority upper limit, the average distance of semantic keywords between the semantic log of environmental influencing factors and the matching environmental semantic log is collected, and the location of environmental influencing factors is performed based on the time series record of the scene where the average distance of semantic keywords is located; Principal component analysis is performed based on the environmental influencing factors of the positioning to achieve optimized positioning. The positioning results are used to connect the relevant mapping table of environmental optimization means, and the environmental optimization means are obtained based on the breeding characteristics of the duck farm.

[0012] Based on the similarity, the optimization of the breeding duck growth scene is performed to obtain multiple priority optimization items. After the optimization items are used to perform the environmental optimization of the duck farm, the following is also included: According to the second aspect of the present invention, the present invention claims protection for an intelligent duck farm environment management optimization system, an intelligent duck farm environment management optimization system, comprising: one or more processors; A memory having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent duck farm environment management optimization method.

[0013] The present application relates to the field of livestock and poultry breeding technology, and in particular to an intelligent duck farm environmental management optimization method and system, which collects real-time growth status information of breeding ducks, uses sensors to optimize the collection of environmental influencing factors of concern to the breeding duck farm, and performs association to obtain growth status information after information synchronization; in the growth status information block, high-weight environmental influencing factors are filtered based on growth attributes, and key environmental influencing factors of multiple growth scenarios are set based on high-weight environmental influencing factors; environmental influencing factors of multiple growth scenarios are collected, and matching environmental logs for healthy growth of breeding ducks are configured to perform semantic processing, and the similarity between the collection and matching environmental logs is collected; based on the similarity, the breeding duck growth scene is optimized, priority optimization items are obtained, and the environmental optimization of the breeding duck farm is performed. The present invention can make optimal value judgments based on the duck breeding status data and the mixed breeding status of breeding ducks in the intelligent duck farm, and effectively improve the utilization efficiency and output results of the duck farm. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A workflow diagram of an intelligent duck farm environment management optimization method claimed in the embodiment of the present application; Figure 2 A second working flow chart of an intelligent duck farm environment management optimization method claimed in the embodiment of the present application; Figure 3 A third work flow chart of an intelligent duck farm environment management optimization method claimed in the embodiment of the present application; Figure 4This is a structural diagram of an intelligent duck farm environment management and optimization system for which protection is sought in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiment is only a structural embodiment of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0016] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second" and "third" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative scene relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a scene, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these scenes, methods, products or devices.

[0017] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present application. The appearance of the phrase in various scenarios in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in a number of other ways described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0019] Figure 1 A flow chart of the intelligent duck farm environment management optimization method of the present invention is shown.

[0020] like Figure 1As shown, the first aspect of the present invention provides an intelligent duck farm environment management optimization method, comprising: A2, collects real-time growth status information of breeding ducks, uses sensors to optimize the collection of environmental factors affecting duck farms, associates historical environmental factors with growth status information, and obtains growth status information after information synchronization; A4, clustering the synchronized growth status information through multiple growth scenarios, filtering high-weight environmental influencing factors based on growth attributes in multiple growth status information blocks, and setting key environmental influencing factors of multiple growth scenarios based on the high-weight environmental influencing factors; A6, using the key environmental influencing factors to collect environmental influencing factors for multiple growth scenarios, and configuring matching environmental logs for healthy growth of breeding ducks, performing semantic processing on the collected environmental influencing factor logs, and collecting similarities with the matching environmental logs; A8, based on the similarity, optimize the breeding duck growth scene to obtain multiple priority optimization items, and use the optimization items to perform environmental optimization of the duck farm.

[0021] It should be noted that the growth information of breeding ducks in duck farms within a preset period is collected, including information such as growth mode and feeding information, the growth information of breeding ducks is subjected to information cleaning and dimensionality reduction and other analyses, and the key scenes of breeding duck growth are collected to perform clustering on the parsed growth information of breeding ducks, and growth information sets of key scenes are collected based on the clustering records, wherein the key scenes look for one or several with longer periods in the pre-production period, production period, post-production period and duckling period; process features are collected based on the growth information set, query ID is configured through the process features to execute query of breeding duck growth samples, breeding duck growth sample records that meet the preset quality are collected based on similarity analysis, and multiple breeding duck growth samples are merged and collected A breeding strategy that integrates real-time growth means; based on the growth information set obtained by clustering driven by the breeding strategy, real-time growth status information of breeding ducks is created, and the growth status information is expanded based on the synchronization of the breeding duck growth information; the environmental data block optimization of the duck breeding farm is performed by collecting and executing the environmental influencing factors of multiple partitions through the GPS positioning system in combination with the wireless sensor element, and the historical environmental influencing factors within a preset period are optimized; the historical environmental influencing factors are associated with the growth status information, and the information influencing factors are set based on the growth status value in the growth status information. The historical environmental optimization information is integrated with the growth status information through the information influencing factors to obtain the growth status information after information synchronization.

[0022] Figure 2 A flow chart of the key environmental influencing factors for setting multiple growth scenes according to the present invention is shown.

[0023] Based on the embodiment of the present invention, high-weight environmental influencing factors are filtered based on growth attributes in multiple growth status information blocks, and key environmental influencing factors of multiple growth scenarios are set based on the high-weight environmental influencing factors, specifically: B2, executing a preset cluster number in the synchronized growth state information through the key scene, presetting a neighborhood range in multiple first cluster number scenes, searching for a growth state value in the neighborhood range, and setting a scene edge based on the mean value corresponding to the growth state value of each cluster after the preset clustering; B4, setting the loss function through the cosine distance, judging the cosine distance of the growth state value in the neighborhood range from the edge of the two adjacent scenes, and when any cosine distance is not less than the preset distance upper limit, the growth state value is associated with the growth scene with high loss; B6, when both cosine distances are not greater than the preset distance upper limit, the growth state value is sequentially associated with the two adjacent growth scenes, and when the clustering of the growth state values ​​in the neighborhood range of all the first cluster number scenes is completed, the cluster record of the synchronized growth state information is collected; B8, collecting growth attributes based on the growth status values ​​in the multiple growth status information blocks, analyzing linear correlation coefficients among environmental influencing factors through the growth attributes, and filtering the environmental influencing factors of the multiple growth status information blocks using the linear correlation coefficients; B10, configuring a breeder duck productivity detection model, training the breeder duck productivity detection model through environmental influencing factors of multiple growth status information blocks, collecting corresponding detection confidences, focusing on maximizing the detection confidences, and evaluating the weights of environmental influencing factors through control experiments; B12, sorting the environmental influencing factors of the multiple growth status information blocks by weight, finding a preset number of high-weight environmental influencing factors and setting them as key environmental influencing factors.

[0024] It should be noted that since the preset clustering of key scenes cannot guarantee the accurate clustering of information, and the structural growth state values ​​are involved in adjacent growth scenes, a neighborhood range is preset in multiple first cluster number scenes. When the growth state value in the neighborhood range meets the hard clustering condition, that is, any distance is not less than the preset distance upper limit, the correlation is analyzed based on the cosine distance. The closer the distance, the greater the correlation, and the growth state value is clustered to the growth scene with a larger correlation; when the growth state value in the domain range meets the soft clustering condition, that is, both cosine distances are not greater than the preset distance upper limit, the growth state values ​​are clustered to the two adjacent growth scenes, so that the clustering of the growth scene is more in line with the actual situation, increasing the confidence of subsequent unexpected optimization.

[0025] The breeding duck productivity detection model is configured based on deep learning methods such as HMM network and Gaussian neural network, with the focus on maximizing the detection confidence. The weighted detection of environmental influencing factors is preferably performed through S control experiments. The environmental influencing factors are evenly divided into s records, each record is set as the experimental set in turn, and the rest is set as the training set. The environmental influencing factors are detected based on the accuracy of the s breeding duck productivity detection models obtained through training.

[0026] Figure 3 A flow chart of the present invention performing semantic processing on an environmental influencing factor log is shown.

[0027] Based on the embodiment of the present invention, semantic processing is performed on the collected environmental influencing factor logs to collect the similarity with the matching environmental logs, specifically: C2, based on key environmental influencing factors, collect environmental influencing factors of multiple growth scenarios, and combine the optimization time to obtain environmental influencing factor logs of multiple growth scenarios; C4, based on the historical average breeding duck productivity, obtain the preset weights of multiple growth scenarios, collect the values ​​and emergency samples of multiple growth scenarios based on the historical breeding duck growth information, and analyze the emergency rates of multiple growth scenarios; C6, using the preset weights in combination with the emergency rate to determine the semantic clustering levels of multiple growth scenarios, clustering the environmental influencing factor logs based on the semantic clustering levels of the multiple growth scenarios, and collecting the corresponding environmental influencing factor semantic logs; C8, and calls the matching environment log to obtain the corresponding matching environment semantic log, analyzes the average distance between the environment influencing factor semantic log and the matching environment semantic log, and obtains the similarity based on the average distance log.

[0028] It should be noted that adaptive semantic clustering levels are set for multiple growth scenarios to effectively capture the changes in influencing factors of environmental influencing factor logs, and semantic clustering levels are obtained based on the values ​​of multiple growth scenarios and emergency samples. If the number of occurrences of diseases, natural disasters, emergencies, etc. in real-time growth scenarios increases, the larger the semantic clustering level, the higher the corresponding clustering density level, thereby achieving high-frequency and refined optimization of growth scenarios.

[0029] The key environmental influencing factors corresponding to multiple growth status information blocks are collected, and the influencing factors are collected in the historical environmental influencing factors based on the key environmental influencing factors, and the influencing factors are filtered based on the corresponding breeding duck productivity in the collected influencing factors; the number of influencing factors under the key environmental influencing factors is collected, and the key environmental influencing factors that are not greater than the preset number are filtered and numbered, and the influencing factors corresponding to the numbered key environmental influencing factors are imported into the servo network to collect candidate production states, and an intrusion strategy is adopted; the sensor structure of the servo network is set to the server to learn the candidate production states of the influencing factors, and the breeding duck growth information of the influencing factors is collected and associated with the same scene. The sensor structure actively learns the candidate production states of the influencing factors, and the candidate features are obtained by sampling. The sampling noise increases the diversity of the obtained samples; a listener is added to the candidate production state layer to prompt the breeding duck growth information of the candidate production state pattern distribution with too large similarity; the influencing factors corresponding to the numbered key environmental influencing factors are expanded through iterative training of the intrusion strategy. When the number of influencing factors meets the preset number, the iteration is stopped, and the influencing factors corresponding to the expanded numbered key environmental influencing factors are output, which solves the problem of decreased optimization confidence caused by information imbalance and improves the robustness of breeding duck growth optimization. The matching environment logs for the healthy growth of breeder ducks in multiple growth scenarios are configured through influencing factors after filtering and information expansion.

[0030] The present invention helps to improve the efficiency, accuracy and value of feeding and breeding traceability, meet the demand for feeding and breeding traceability, and realize a more effective feeding and breeding marketing path.

[0031] Feed management: For both laying ducks and meat ducks, the quality and type of feed are key factors affecting their production performance. Laying ducks require high-quality protein and energy during the laying period to ensure egg production and egg quality. Meat ducks require high-energy, high-protein feed to promote rapid growth. In addition, the feed formula should be adjusted according to the different growth stages of ducks, such as duckling stage, young duck stage and adult duck stage, to ensure a balanced supply of nutrition.

[0032] Rearing environment management: Appropriate temperature, humidity and density are essential for the healthy growth of ducks. Ducklings are sensitive to temperature and need to maintain a suitable temperature to avoid problems caused by low or high temperatures. Humidity and density also need to be adjusted according to the season and the growth stage of the duck to provide a comfortable growth environment. Disease prevention and control: Ducks are more sensitive to diseases, so it is very important to prevent and control the occurrence of diseases. This includes regularly checking the health of the ducks, isolating sick ducks in time, and keeping the breeding environment clean and hygienic. In addition, for egg-laying ducks, it is also very important to keep the environment quiet, because ducks are sensitive to environmental changes, and excessive stress may lead to a decrease in egg production.

[0033] In one embodiment, detection and information extraction are performed on breeding ducks, and breeding information is output, including: Perform purpose detection on breeding ducks to determine the purpose of breeding ducks. When the breeding ducks are raised for eggs, extract the lighting information and feeding space information of the breeding ducks, perform lighting quality mining on the lighting information, output the lighting quality information, extract the feeding space information as the corresponding feeding density information based on the configured feeding database, perform collaborative operations on the feeding density information and the lighting quality information, and output the feeding information; When the breeding ducks are for feather breeding, the breed information, feather grade information, chromaticity information and feather stocking density information of the breeding ducks are extracted, the breed information and feather grade information are subjected to information extraction according to the configured feather strategy, and the feather information is output, and the feather stocking density information and chromaticity information are sequentially extracted through the breeding database, and the feather stocking density information and chromaticity information are sequentially output, and the chromaticity information and the stocking density information are attached and coordinated to the feather information, and the breeding information is output; When breeding ducks are raised for meat, the meat breeding data log, feed attribute information, disease prevention information and meat texture video of the breeding ducks are extracted, meat quality information mining is performed on the extracted feed attribute information, disease prevention information and meat texture video, and feed quality information, disease prevention quality information and product display quality information are output. The collaborative breeding database extracts the meat breeding data log as meat breeding density information, performs collaborative packaging operations on the meat breeding density information, feed quality information, disease prevention quality information and product display quality information, and outputs breeding information.

[0034] This embodiment is based on the purpose detection and information extraction of breeding ducks, which can more accurately understand the breeding content and characteristics, and help to accurately locate the target customer group. Performing different operations on breeding ducks for different purposes can improve the relevance and attractiveness of the breeding content, thereby enhancing the value of breeding and the attractiveness of breeding ducks. Based on operations such as light quality mining, variety information extraction, and stocking density information extraction, breeding can be made to better meet the interests and needs of breeding ducks and improve the breeding duck experience. Based on the breeding information output by the mining operation, more information support and decision-making coordination are provided for breeding traceability, which helps to optimize the breeding traceability path and improve the conversion rate of breeding. Automating the processes of blockchain compression information decompression, breeding duck detection, and information extraction can improve the synchronization and mining efficiency of breeding information and quickly optimize the breeding traceability path.

[0035] It should be noted that the preset similarity upper limit range associates optimization priorities with multiple preset similarity upper limit ranges, and the corresponding optimization priority is determined by the similarity upper limit range in which the collected similarity falls, and the optimization items are transmitted and listed according to the preset method; when the optimization priority is greater than the preset priority upper limit, the average distance of the semantic keywords between the semantic log of the environmental influencing factor and the matching environmental semantic log is collected, and the environmental influencing factor positioning is performed based on the time series record of the scene where the average distance of the semantic keywords is located; principal component analysis is performed based on the located environmental influencing factors to achieve optimized positioning, and the positioning results are used to connect the relevant mapping table of environmental optimization means, and the environmental optimization means are obtained based on the attention to the breeding characteristics of duck farms.

[0036] Furthermore, the mixed breeding status of breed ducks of each duck farm is collected according to the duck farm management server, and the growth progress deviation value of each breed duck is analyzed according to the mixed breeding status of breed ducks; If the growth progress deviation value of the breeder ducks is greater than the normal growth progress deviation value, collecting the real-time video mixed breeding status of the duck farm, and collecting the video recognition result based on the real-time video mixed breeding status of the duck farm; Collecting the key growth time nodes of each breed of ducks, and generating the market value effect of each duck farm according to the video recognition results and the key growth time nodes of each breed of ducks; The rearing environment of each duck farm is adjusted based on the market value of each duck farm, and the quality status of the feed in the duck farm is collected, and prompts are given according to the quality status of the feed in the duck farm.

[0037] Furthermore, the mixed breeding status of breed ducks of each duck farm is collected according to the duck farm management server, and the growth progress deviation value of each breed duck is analyzed according to the mixed breeding status of breed ducks, specifically: The low-value mixed breeding status in the duck farm is collected according to the historical breeding records, and the species information of the breed ducks in the duck farm is determined according to the low-value mixed breeding status in the duck farm, and the query tag is constructed based on the species information of the breed ducks in the duck farm; The breed type of each breed of duck is collected according to the query tag, and the relevant environmental type monitoring equipment is deployed according to the breed type of each breed of duck, and a breeding monitoring platform is constructed based on the relevant environmental type monitoring equipment; The mixed breeding status of breed ducks in each duck farm is collected through the breeding monitoring platform, and the matching breed duck data of each breed duck is collected according to the historical breeding records to build a growth progress detection model for breed ducks; The growth prospects of breed ducks are preset, and the mixed breeding status of breed ducks in each duck farm, the matching breed duck data of each breed duck, and the growth prospects of breed ducks are input into the growth progress detection model of breed ducks to generate the growth progress deviation value of each breed duck in each duck farm.

[0038] Among them, in this embodiment, the video recognition result is collected according to the video mixed breeding status of the real-time duck farm, specifically: A breeding duck breed recognition model is constructed, and a large number of breeding duck mixed breeding status videos are collected based on historical breeding records, and a decision tree algorithm is introduced to extract breeding duck features from the breeding duck mixed breeding status videos through the decision tree algorithm to generate frame sequences of various duck species; Inputting the frame sequences of the various duck species into the breeding duck variety identification model for coding learning, and outputting the breeding duck variety identification model when the model parameters of the breeding duck variety identification model meet the preset requirements; After preprocessing the video polyculture state of the real-time duck farm, a video frame sequence of the real-time duck farm is generated, and the video frame sequence of the real-time duck farm is input into the breeding duck variety recognition model for recognition; After the recognition, the breeding duck breed recognition result in the real-time duck farm video is collected, and the breeding duck breed recognition result in the real-time duck farm video is output as the video recognition result.

[0039] Based on the market value of each duck farm, the rearing environment of each duck farm is adjusted, including: When the market value effect of the duck farm is greater than the preset market value effect, the isolation means of the growing breed ducks are collected according to the historical breeding records, and a breeding duck isolation mapping table of the growing breed ducks is constructed; Analyze the attention score of the isolation means of each growing breed duck, and input the isolation means of growing breed ducks with the same attention score into the same space of the breeding duck isolation mapping table of growing breed ducks; Inputting isolation means of growing breed ducks with different attention scores into different spaces of a growing breed duck isolation mapping table, and collecting isolation means of growing breed ducks of each duck farm through the growing breed duck isolation mapping table, and at the same time, collecting communication protocol information of the environment adjustment equipment of the duck farm according to the duck farm management server; The communication protocol information of the duck farm's environmental adjustment equipment is sent to the remote control terminal, and the remote control terminal is used to control the duck farm's environmental adjustment equipment to adjust the environment and isolate the growing breed ducks of each duck farm.

[0040] Figure 4 A block diagram of the intelligent duck farm environment management optimization system of the present invention is shown.

[0041] According to the second embodiment of the present invention, the present invention claims protection for an intelligent duck farm environment management optimization system, referring to Figure 4 , an intelligent duck farm environment management optimization system, including: one or more processors; A memory having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent duck farm environment management optimization method.

[0042] The third embodiment of the present invention also provides an analysis machine readable storage medium, which includes an intelligent duck farm environment management optimization method program. When the intelligent duck farm environment management optimization method program is executed by a processor, the steps of the intelligent duck farm environment management optimization method as described in any one of the above items are implemented.

[0043] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the clustering of units is only a kind of logical function clustering. There may be other clustering methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the recorded or discussed mutual coupling or direct coupling or communication connection can be an indirect coupling or communication connection of some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0044] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can be physically provided separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. All equivalent structures or equivalent process transformations made through the description and drawings of this application, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

[0045] The above detailed description of the specific implementation of the invention is only set as an example, and the present application is not limited to the specific implementation described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application. Therefore, the equalization, modification, improvement, etc. made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. An intelligent duck farm environment management optimization method, characterized in that: The following steps are involved: Collect the real-time growth status information of the breeding ducks, use sensors to optimize the collection of environmental factors affecting the duck farm, and associate the historical environmental factors with the growth status information to obtain the growth status information after information synchronization; Clustering the synchronized growth status information through multiple growth scenarios, filtering high-weight environmental influencing factors based on growth attributes in multiple growth status information blocks, and setting key environmental influencing factors of multiple growth scenarios based on the high-weight environmental influencing factors; The key environmental influencing factors are used to collect environmental influencing factors of multiple growth scenarios, and matching environmental logs for healthy growth of breeding ducks are configured, semantic processing is performed on the collected environmental influencing factor logs, and the similarity between the collected and matching environmental logs is calculated; Based on the similarity, the growth scenario of the breeding ducks is optimized to obtain multiple priority optimization items, and the optimization items are used to perform environmental optimization of the duck farm.

2. An intelligent duck farm environment management optimization method as claimed in claim 1, characterized in that: Based on the similarity, the optimization of the breeding duck growth scene is performed to obtain multiple priority optimization items. After the optimization items are used to perform the environmental optimization of the duck farm, the following is also included: The mixed breeding status of breed ducks of each duck farm is collected according to the duck farm management server, and the growth progress deviation value of each breed duck is analyzed according to the mixed breeding status of breed ducks; If the growth progress deviation value of the breeder ducks is greater than the normal growth progress deviation value, collecting the real-time video mixed breeding status of the duck farm, and collecting the video recognition result based on the real-time video mixed breeding status of the duck farm; Collecting the key growth time nodes of each breed of ducks, and generating the market value effect of each duck farm according to the video recognition results and the key growth time nodes of each breed of ducks; The rearing environment of each duck farm is adjusted based on the market value of each duck farm, and the quality status of the feed in the duck farm is collected, and prompts are given according to the quality status of the feed in the duck farm.

3. An intelligent duck farm environment management optimization method as claimed in claim 2, characterized in that: The duck farm management server collects the mixed breeding status of breed ducks in each duck farm, and analyzes the growth progress deviation value of each breed duck according to the mixed breeding status of breed ducks, specifically: The low-value mixed breeding status in the duck farm is collected according to the historical breeding records, and the species information of the breed ducks in the duck farm is determined according to the low-value mixed breeding status in the duck farm, and the query tag is constructed based on the species information of the breed ducks in the duck farm; The breed type of each breed of duck is collected according to the query tag, and the relevant environmental type monitoring equipment is deployed according to the breed type of each breed of duck, and a breeding monitoring platform is constructed based on the relevant environmental type monitoring equipment; The mixed breeding status of breed ducks in each duck farm is collected through the breeding monitoring platform, and the matching breed duck data of each breed duck is collected according to the historical breeding records to build a growth progress detection model for breed ducks; The growth prospects of breed ducks are preset, and the mixed breeding status of breed ducks in each duck farm, the matching breed duck data of each breed duck, and the growth prospects of breed ducks are input into the growth progress detection model of breed ducks to generate the growth progress deviation value of each breed duck in each duck farm.

4. The intelligent duck farm environment management optimization method according to claim 1, characterized in that: Based on the historical environmental influencing factors and the growth status information, the growth status information after information synchronization is obtained, which is specifically: Collecting the growth information of the breeding ducks of the duck farm concerned within a preset period, parsing the growth information of the breeding ducks, collecting key scenes of the growth of the breeding ducks, clustering the parsed growth information of the breeding ducks, and collecting the growth information set of the key scenes based on the clustering records; Based on the growth information set collection process characteristics, query the growth samples of breeding ducks by configuring the query ID through the process characteristics, collect the growth sample records of breeding ducks that meet the preset quality based on similarity analysis, and merge and collect the breeding strategies of real-time growth means for multiple breeding duck growth samples; Creating real-time growth status information of the breeding ducks based on the growth information set obtained by clustering driven by the breeding strategy, and performing expansion of the growth status information based on the synchronization of the breeding ducks growth information; Through the GPS positioning system combined with wireless sensor elements, the environmental data block optimization of the duck farm is carried out to optimize the environmental influencing factors of multiple partitions and collect historical environmental influencing factors within a preset period; The historical environment influencing factors are associated with the growth status information, the information influencing factors are set based on the growth status value in the growth status information, and the historical environment optimization information is fused with the growth status information through the information influencing factors to obtain the growth status information after information synchronization.

5. The intelligent duck farm environment management optimization method according to claim 1, characterized in that: In multiple growth status information blocks, high-weight environmental influencing factors are filtered based on growth attributes, and key environmental influencing factors of multiple growth scenarios are set based on the high-weight environmental influencing factors, specifically: By executing a preset cluster number in the synchronized growth state information of the key scene, a neighborhood range is preset in multiple first cluster number scenes, the growth state value in the neighborhood range is searched, and the scene edge is set based on the mean value corresponding to the growth state value of each cluster after the preset clustering; The loss function is set by cosine distance, and the cosine distance of the growth state value in the neighborhood range from the edge of the two adjacent scenes is determined. When any cosine distance is not less than the preset distance upper limit, the growth state value is associated with the growth scene with high loss. When both cosine distances are not greater than the preset distance upper limit, the growth state value is sequentially associated with the two adjacent growth scenes. When the clustering of the growth state values ​​in the neighborhood range of all the first cluster number scenes is completed, the cluster record of the synchronized growth state information is collected; Collecting growth attributes based on growth status values ​​in multiple growth status information blocks, analyzing linear correlation coefficients in environmental influencing factors through the growth attributes, and filtering environmental influencing factors of multiple growth status information blocks using the linear correlation coefficients; A breeding duck productivity detection model is configured, the breeding duck productivity detection model is trained by environmental influencing factors of multiple growth status information blocks, corresponding detection confidences are collected, maximizing the detection confidences is focused on, and the weights of environmental influencing factors are evaluated through control experiments; The environmental influencing factors of the plurality of growth status information blocks are sorted by weight, and a preset number of high-weight environmental influencing factors are found and set as key environmental influencing factors.

6. The intelligent duck farm environment management optimization method according to claim 1, characterized in that: The key environmental influencing factors are used to collect environmental influencing factors for multiple growth scenarios, and a matching environmental log for the healthy growth of breeding ducks is configured, specifically: Collecting key environmental influencing factors corresponding to a plurality of growth status information blocks, collecting influencing factors from historical environmental influencing factors based on the key environmental influencing factors, and filtering influencing factors based on corresponding breeding duck productivity among the collected influencing factors; Collect the number of influencing factors under key environmental influencing factors, filter the execution number of key environmental influencing factors that are not greater than the preset number, import the influencing factors corresponding to the numbered key environmental influencing factors into the servo network to collect candidate production states, and adopt the intrusion strategy; The sensor structure of the servo network is set to the candidate production state of the server learning influencing factors, the growth information of the breeding ducks of the influencing factors is collected and associated with the same scene, and a listener is added to the candidate production state layer to prompt the growth information of the breeding ducks with too large distribution similarity of the candidate production state patterns; The influencing factors corresponding to the numbered key environmental influencing factors are expanded through iterative training of the intrusion strategy. When the number of influencing factors meets the preset number, the iteration is stopped, and the influencing factors corresponding to the expanded numbered key environmental influencing factors are output; The matching environment logs for the healthy growth of breeder ducks in multiple growth scenarios are configured through influencing factors after filtering and information expansion.

7. The intelligent duck farm environment management optimization method according to claim 1, characterized in that: Perform semantic processing on the collected environmental influencing factor logs and collect the similarity with the matching environmental logs, specifically: Based on the key environmental influencing factors, the environmental influencing factors of multiple growth scenarios are collected, and the environmental influencing factor logs of multiple growth scenarios are obtained by combining the optimized time; Based on the historical average breeding duck productivity, the preset weights of multiple growth scenarios are obtained, and based on the historical breeding duck growth information, the values ​​and emergency samples of multiple growth scenarios are collected to analyze the emergency rates of multiple growth scenarios; Determine the semantic clustering levels of multiple growth scenarios by using the preset weights in combination with the emergency rate, perform clustering of environmental influencing factor logs based on the semantic clustering levels of multiple growth scenarios, and collect corresponding environmental influencing factor semantic logs; The matching environment log is called to obtain the corresponding matching environment semantic log, the average distance between the environment influencing factor semantic log and the matching environment semantic log is analyzed, and the similarity is obtained based on the average distance log.

8. The intelligent duck farm environment management optimization method according to claim 1, characterized in that: Based on the similarity, the optimization of the breeding duck growth scene is performed to obtain multiple priority optimization items, and the optimization items are used to perform the environmental optimization of the duck farm, specifically: Preset similarity upper limit range, associate optimization priorities for multiple preset similarity upper limit ranges, determine the corresponding optimization priority for the similarity upper limit range in which the collected similarities fall, and obtain the optimization items to be transmitted and recorded in a list according to the preset method; When the optimization priority is greater than the preset priority upper limit, the average distance of semantic keywords between the semantic log of environmental influencing factors and the matching environmental semantic log is collected, and the location of environmental influencing factors is performed based on the time series record of the scene where the average distance of semantic keywords is located; Principal component analysis is performed based on the environmental influencing factors of the positioning to achieve optimized positioning. The positioning results are used to connect the relevant mapping table of environmental optimization means, and the environmental optimization means are obtained based on the breeding characteristics of the duck farm.

9. An intelligent duck farm environment management optimization system, characterized in that: include: one or more processors; A memory having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement an intelligent duck farm environment management optimization method according to any one of claims 1 to 8.

Images